November 27, Financial Accounting Standards Board 401 Merritt 7 P.O. Box 5116 Norwalk, CT

Transcription

1 November 27, 2013 Financial Accounting Standards Board 401 Merritt 7 P.O. Box 5116 Norwalk, CT Exposure Draft Insurance Contracts File Reference No The Financial Reporting Executive Committee (FinREC) and the Insurance Expert Panel, both of the American Institute of Certified Public Accountants (AICPA), appreciate the opportunity to comment on the Financial Accounting Standards Board (FASB) Exposure Draft Insurance Contracts ( FASB ED ). Additional input was also received by members of the AICPA Depository Institutions Expert Panel, Employee Benefit Plans Expert Panel and Health Care Expert Panel. FinREC continues to support the efforts of both the FASB and the International Accounting Standards Board (IASB) to either converge or more closely align U.S. generally accepted accounting principles (U.S. GAAP) with the International Financial Reporting Standards (IFRS). However, as we have stated in previous letters, high quality accounting standards should not be sacrificed for the sake of convergence. We also note that U.S. generally accepted accounting principles (GAAP) comprehensively addresses accounting for insurance contracts by insurance entities, whereas IFRSs do not have comprehensive insurance guidance. We understand that FASB has not yet evaluated if the benefits that are expected from the proposed changes exceed the cost, nor whether targeted improvements should instead be made to U.S. GAAP. Our letter has not addressed areas where targeted improvements, if the FASB decides to follow that approach, should be considered. We believe that if the FASB decides that changes are necessary to U.S. GAAP beyond targeted improvements then the following are key areas that the FASB and the IASB should seek convergence on: Unlocking the Margin - changes in estimates of future cash flows, which are related to future coverages or services, should be recognized as adjustments to the margin (see our response to Question 13) Fulfillment Cash Flows should include a measurement for the uncertainty related to the allocation of probable outcomes (see our response to Question 12) Definition of Portfolio - the definition of a portfolio of insurance contracts should take into account how entities manage their business (see our response to Question 8) Acquisition Costs - what qualifies as acquisition costs and the accounting for acquisition costs paid (see our responses to Questions 28 and 29)

2 Transition practical expedients should allow for the use of hindsight (see our response to Question 44) Our comments in this letter have been prepared to provide feedback on the FASB ED. We do not fully support the building block approach ( BBA ) or the premium allocation approach ( PAA ) as currently proposed or the proposed presentation of the models. Included in our response to the specific questions we have provided suggestions on certain aspects of the proposed approaches. We would suggest that the FASB consider the following observations: Scope: We agree that the guidance should apply to contracts that meet the definition of an insurance contract, and should not be based on the legal type of entity that may have issued the contract. However, we are concerned that without modifications to the definition of an insurance contract the population of arrangements that would be required to use the guidance would be too broad. One area of concern is that the proposed scope includes arrangements that are predominantly based on credit risk as insurance contracts. We believe that the FASB should rethink how arrangements that are predominantly based on credit risk be classified. We believe that contracts whose primary purpose is to provide the issuing entity with exposures to credit risk, that are not currently accounted for as insurance contracts, should be included in the financial instruments project. We also recommend that the FASB consider if changing accounting models for arrangements that are predominantly based on credit risk and are not currently accounted for as insurance contracts, would be appropriate from a cost benefit perspective. Another area of concern is that the FASB ED indicates insurance must cover a pre-existing risk, and this concept is not well understood outside the insurance industry. We believe further clarification is needed to explain if a risk should or should not be considered a preexisting risk and separated from an overall arrangement between two parties for the purposes of assessing whether it is to be considered within the scope of the FASB ED. We are also concerned with the lack of clarity in determining whether fixed-fee service contracts have the primary purpose of providing a service. While we agree with the scope exclusion for fixed-fee service contracts provided in of the FASB ED, we request clarification in determining at what point a contract has the primary purpose of providing a service. Therefore, we recommend that the FASB reevaluate whether the definition of an insurance contract should be modified to avoid expanding the application of the guidance to contracts that should not be accounted for under the insurance contract guidance. Premium Allocation Approach: We recommend that for contracts that would be required to apply the premium allocation approach ( PAA ), but that are managed with contracts accounted for under the BBA, entities should be permitted to apply the BBA to all the related contracts. Some insurance entities may wish to apply the BBA to all insurance contracts managed together for ease of administration and consistency in presentation. Entities would also be required to disclose this election. 2

3 Portfolios: Currently there is diversity in U.S. GAAP with respect to how entities aggregate contracts for measurement resulting in difficulty for users in comparing financial statements. While we agree that the definition of a portfolio of insurance contracts should be addressed in the final guidance, we do not agree with the definition included in the FASB ED. We strongly recommend that the FASB and the IASB converge on this issue, as differences in the definition of a portfolio of insurance contracts will result in significant reporting complications for multinational entities. The definition of a portfolio of insurance contracts as defined in the IASB exposure draft ( IASB ED ), that takes into account how entities manage their business, is more appropriate. The proposed definition of an overall portfolio of insurance contracts in the FASB ED may require a more granular level of portfolios than how entities manage their business and may not be justifiable from a cost benefit perspective. We also believe that for certain measurement amounts it may be necessary to group insurance contracts into a smaller unit of account than the portfolio. For example, when determining the discount rate and margin, and unlocking the margin. We recommend that the final guidance acknowledge that different groupings would be permitted. Discount Rates: We agree that an entity should separately present the effects of underwriting performance from the effects of changes in discount rates, but do not agree that changes in the present value of the fulfillment cash flows due to changes in the discount rates should be required to be included in other comprehensive income due to the potential accounting mismatches and resulting volatility in earnings. We believe that an entity should be allowed to make an accounting policy decision in an attempt to mitigate volatility in earnings (similar to the fair value option) in regards to whether changes in discount rates should be recognized in other comprehensive income or net income. We believe that this election should be consistent with any final decisions on the fair value option election under the FASB s financial instruments project. Fulfillment Cash Flows: We are concerned that using fulfillment cash flows (the present value of the explicit, unbiased and probability-weighted estimates of the future cash flows) as defined in the FASB ED would not include a measurement of the uncertainty related to the allocation of probable outcomes, and may not accurately reflect the measurement of the insurance liability. We believe that both the BBA and PAA should include the principle that there is uncertainty in cash flows related to the allocation of probable outcomes. There is diversity among the various individuals within the AICPA group that compiled this letter regarding how the uncertainty should be included in the measurement models. There are various ways that uncertainty could be captured in both approaches such as; including an explicit or implicit risk adjustment or amortization of the margin or premium over the coverage and settlement period. We also request further clarification as to the unbiased measurement of cash flows and the interaction with probability-weighted estimates. It could be inferred that the use of 3

4 probability-weighted estimates implicitly includes the use of management bias, to determine the weightings. We recommend that the final guidance further elaborate on what judgments are allowable and still considered unbiased. Specifically we believe uncertainty is not fully included in the PAA, as the premium is amortized only over the coverage period and the liability for incurred claims is discounted. We recommend that the FASB reconsider how to include the uncertainty in the cash flows. Unlocking of the Margin: We believe that changes in estimates of future cash flows, which are related to future coverages or services, should be recognized as adjustments to the margin instead of in net income. The original amount included in the margin is an estimate of future expectations, and we do not believe it is appropriate to recognize current period refinements of the estimate for those future cash flows in net income. We believe recognizing estimates of future cash flows and changes to future expectations of those estimates both in the margin will result in more relevant measure of unearned profit in the financial statements. Acquisition Costs: We request that the FASB and the IASB strive for convergence on what qualifies as acquisition costs, as we believe it will be confusing to users of financial statements. We believe that costs for unsuccessful efforts should not be included as acquisition costs. We also request that the Boards strive for convergence on accounting for qualifying acquisition costs as a component of the margin or included as part of fulfillment cash flows. We believe that this is a fundamental difference that will make comparability among U.S. and international insurance entities unnecessarily difficult. Business Combinations: We do not agree that entities should record a loss at the acquisition date in the amount by which any excess of the asset and liability balances related to insurance contracts measured in accordance with the guidance in the FASB ED exceeds the fair value of those assets and liabilities. We believe that the proposal in the FASB ED to recognize a loss at the acquisition date would be an overall exception to the principles in ASC 805, Business Combinations, that has not been thoroughly explained. If the FASB proceeds with the guidance proposed, we request further elaboration why such an exception would be appropriate for the acquisition of insurance contracts under a business combination. Presentation: Although we believe it is better to present insurance contract revenue and incurred expenses in the statement of comprehensive income rather than only changes in the margin, we are concerned that the current proposed presentation approach may not provide users with a relevant financial measure. 4

5 We agree with the objective of providing volume information in the statement of comprehensive income for contracts under both the BBA and the PAA that aligns the underlying concepts with the principles outlined in the anticipated revenue recognition standard. However for insurance contracts under the BBA the period of time from inception until settlement can be for many years, and often decades, so following the principles under the proposed revenue recognition standard may not meet user needs and would require extensive system enhancements to capture the necessary data. Therefore, we believe further outreach is needed to make sure the information included is helpful to users of insurance entity financial statements before insurers are required to expend significant cost to comply with the requirements of the FASB ED. We also believe that the FASB should reconsider presentation for reinsurance arrangements. Under the principles in in FASB ASC , Asset Retirement and Environmental Obligations (SOP 96-1, Environmental Remediation Liabilities), and FASB ASC , Revenue Recognition Gains and Losses (FASB Interpretation 30, Accounting for Involuntary Conversions of Nonmonetary Assets to Monetary Assets), related to presenting insurance recoveries in the same income statement line, we believe that recoveries pertaining to reinsurance contracts should be allowed to be presented net in direct losses on the statement of comprehensive income. Consistent with this concept, we also believe that premiums for reinsurance arrangements should be presented net, and disclosed in the notes to the financial statements by direct, assumed and ceded. We believe this presentation is appropriate as it provides a better understanding of the overall financial statements. We accept that permitting a net presentation may not entirely align other accounting guidance, but all the necessary details could be provided in the footnotes. We are also concerned that presenting portfolios of insurance contracts separately as net insurance contract liabilities and net insurance contract assets would be very confusing to users of insurance entity financial statements. We recommend that all portfolios of insurance contracts be presented together, either as an asset or liability, and that the asset and liability positions be further detailed in the notes to the financial statements. We recommend that during the upcoming roundtables, the FASB reach out to financial statement users to gather feedback on whether the proposed information is useful, and what other information could be displayed. Transition: We support practical expedients being provided for transition, but have concerns that some of the practical expedients in the FASB ED will still require extensive work to obtain older information. Conceptually we do not believe that it is appropriate to require the margin at transition to be zero if it is impracticable to apply the guidance retrospectively or if there is no objective information that is reasonably available. We believe that the practical expedients provided in the IASB ED are more appropriate in allowing for hindsight. We believe allowing entities to use a modified retrospective application as described in the IASB ED (allowing for the use of known activity to approximate historical information when it is impracticable to obtain), would allow for more consistent information and comparability 5

6 among financial statements as this will result in less situations with portfolios of insurance contracts with zero margins, and more verifiability of inputs. We believe that the requirement in (j) of the FASB ED, for business combinations that occurred before the transition date, to determine the margin as of the original acquisition date by comparing the fair value of the asset and liability balances related to insurance contracts to the expected fulfillment cash flows may not be operational due to limitations on available past information. We recommend that the FASB consider allowing a practical expedient to permit entities to use hindsight in determining the expected fulfillment cash flows at the date of acquisition. Our answers to the specific questions in the FASB ED provide more detail on the views expressed above and are attached in the Appendix to this letter. We have also attached as reference our comment letter to the IASB as an appendix to this letter. Yours truly, Richard Paul, Chair Financial Reporting Executive Committee Richard Lynch, Chair ( ) Insurance Expert Panel Richard Sojkowski, Chair ( ) Insurance Expert Panel 6

7

8

9 Appendix A Response to Questions: FASB Exposure Draft: Insurance Contracts Scope Question 1: Do you agree with the scope and the scope exclusions included in this proposed guidance, including its applicability to contracts written by noninsurance entities? If not, what types of contracts or transactions also should be included or excluded from the scope and why? Yes, we agree that the guidance should apply to contracts that meet the definition of an insurance contract, and should not be based on the legal type of entity that may have issued the contract. However we are concerned that without modifications to the definition of an insurance contract, the population of arrangements that would be required to use the guidance would be too broad. One area of concern is that the proposed scope includes arrangements that are predominantly based on credit risk as insurance contracts. We believe that the FASB should rethink how arrangements that are predominantly based on credit risk (for example standby letters of credit) be classified. We believe that contracts whose primary purpose is to provide the issuing entity with exposures to credit, that are not currently accounted for as insurance contracts, should be included in the financial instruments project. Including these products in the financial instruments project will allow for comparability of credit risk-related products within the banking industry. We also recommend that the FASB consider if changing accounting models for arrangements that are predominantly based on credit risk and are not currently accounted for as insurance contracts, would be appropriate from a cost benefit perspective. Another area of concern is that the FASB ED indicates insurance must cover a pre-existing risk, and this concept is not well understood outside the insurance industry. We believe further clarification is needed to explain if a risk should or should not be considered a preexisting risk and separated from an overall arrangement between two parties for the of assessing whether it is to be considered within the scope of the FASB ED. We are also concerned with the lack of clarity in determining whether fixed-fee service contracts have the primary purpose of providing a service. While we agree with the scope exclusion for fixed-fee service contracts provided in of the FASB ED, we request clarification in determining at what point a contract has the primary purpose of providing a service. Therefore we recommend that the FASB reevaluate whether the definition of an insurance contract should be modified to avoid expanding the application of the guidance to contracts that should not be accounted for under the insurance contract guidance. The FASB ED excludes retirement benefit obligations reported by defined benefit retirement plans that follow the guidance in FASB Topic 960. We recommend that this scope exception extend to health and welfare benefit plans (as defined within Topic 965), as these plans offer

10 postretirement and postemployment benefits as well as employer provided insurance benefits that are similar to insurance contracts. If these contracts are not excluded from the scope of the FASB ED, there would potentially be conflicting guidance for these plans to follow, as a health and welfare benefit plan would first look to Topic 965 for assistance. We also request that the final guidance, in either the basis for conclusion or an example in the implementation guidance, clarify why defined contribution plans (as defined within Topic 962) are not specifically excluded from the scope. The benefits provided by defined contribution plans are limited to the balances accumulated in the individual participant accounts, with no further obligation or promise from the employer, and therefore would not need to be evaluated for significant insurance risk. We strongly support the inclusion of examples in the implementation guidance of what products with guarantees meet or do not meet the definition of insurance contracts ( of the FASB ED). We believe the current examples provide insufficient background about the arrangement to understand how the FASB decided whether or not the arrangement met the definition on an insurance contract or met the definition but was specifically scoped out. Therefore, we request more clarification of the descriptions of products (arrangement) and what obligates (e.g. own performance) the one party to compensate the other party. We are concerned that some of the answers on the examples provided in the FASB ED appear to be inconsistent with the framework of the proposed guidance. Members of the AICPA Depository Expert Panel and the Insurance Expert Panel would be happy to work with the FASB to enhance the examples to be included in the final guidance. We also recommend that one additional column should be added to the chart in of the FASB ED, to explain why the example meets or does not meet the criteria to be insurance. The way the chart is currently drafted it is unclear as to whether all of the scope exceptions have been considered in determining whether the examples are to be included in the scope. Recognition Question 2: Do you agree with the requirements included in this proposed Update for when noninsurance components of an insurance contract, including embedded derivatives, distinct investment components, and distinct performance obligations to provide goods or services, should be separately accounted for under other applicable Topics? If not, why? Yes, we agree that is appropriate to separate certain components of an insurance contract when the noninsurance components are clearly separable from the insurance component. Under the proposed guidance we are unclear and request clarification as to whether a distinct performance obligation to provide services (such as investment services), related to an investment component that is not separated from the insurance component should be separated. 2

11 We support the inclusion of examples ( in the FASB ED) of how to apply the proposed guidance for separation of goods and services, but recommend that the final guidance include additional more realistic examples such as asset management services and roadside assistance coverage bundled with the sale of a car. We also request further explanation why the Stop-Loss and High Deductible Health Insurance Plan examples result in different conclusions when claims processing services are sold separately for both products. Measurement Approaches Question 5: Do you agree that entities should apply different approaches to contracts with different characteristics, described as the building block approach and the premium allocation approach? If not, which model do you think should apply and do you think there should be any changes made to that model? Yes, we agree in theory that entities should apply different approaches (the building block and the premium allocation) to contracts with different characteristics. However, we recommend that for contracts that would be required to apply the PAA, but that are managed with contracts accounted for under the BBA, entities should be permitted to apply the BBA to all the related contracts. Some insurance entities may wish to apply the BBA to all insurance contracts managed together for ease of administration and consistency in presentation. Entities would also be required to disclose this election. Question 6: Do you agree that entities should be required to apply the premium allocation approach if the coverage period of the insurance contract, considering the contract boundary guidance, is one year or less? If not, what would you recommend and why? As noted above in our response to Question 5, we believe that entities should be permitted to apply the BBA to all related contracts that are managed together for ease of administration and consistency in presentation. Question 7: Do you agree that entities should be required to apply the premium allocation approach if at contract inception, it is unlikely that during the period before a claim is incurred there will be significant variability in the expected value of the net cash flows required to fulfill the contract? If not, what do you recommend and why? As noted above in our response to Question 5, we believe that entities should be permitted to apply the BBA to all related contracts that are managed together for ease of administration and consistency in presentation. We request that the FASB provide clarification as to whether the assessment for significant variability in the expected value of the net cash flows required to fulfill the contract should be made on a nominal basis or present value basis. The proposed guidance appears to use different definitions of expected value throughout the document. 3

12 We also request clarification as to what level should the determination of not expecting significant variability be made, the contract or portfolio level? of the FASB ED defines expected value as the present value of the unbiased, probability-weighted estimate of the future cash outflows less the future cash inflows. While of the FASB ED discusses that the expected value is that statistical mean of the full range of possible outcomes. We request clarification as to whether the expected value inherently includes the time value of money, or if that is a separate step to consider after the expected value is determined. Question 8: Do you agree with definition of a portfolio of insurance contracts as included in this proposed guidance? If not, what do you recommend and why? Currently there is diversity in U.S. GAAP with respect to how entities aggregate contracts for measurement resulting in difficulty for users in comparing financial statements. While we agree that the definition of a portfolio of insurance contracts should be addressed in the final guidance, we do not agree with the definition included in the FASB ED. We strongly recommend that the FASB and IASB converge on this issue, as differences in the definition of a portfolio of insurance contracts will result in significant reporting complications for multinational entities. The definition of a portfolio of insurance contracts as defined in the IASB ED, that takes into account how entities manage their business, is more appropriate. The proposed definition of an overall portfolio of insurance contracts in the FASB ED may require a more granular level of portfolios than how entities manage their business and may not be justifiable from a cost benefit perspective. We also believe that for certain measurement amounts it may be necessary to group insurance contracts into a smaller unit of account than the portfolio. For example, when determining the discount rate and margin, and unlocking the margin. We recommend that the final guidance acknowledge that different groupings would be permitted. Fulfillment Cash Flows Question 10: Do you agree with the types of cash flows that would be included in the measurement of the fulfillment cash flows, including embedded options and guarantees related to the insurance coverage under the existing insurance contract that are not separated and accounted for as embedded derivatives? If not, what cash flows do you think also should be included or excluded and why? We also request that the FASB and the IASB strive for convergence on accounting for qualifying acquisition costs as a component of the margin or included as part of fulfillment cash flows. We believe that this is a fundamental difference that will make comparability among U.S. and international insurance entities unnecessarily difficult. In determining what should be included as part of fulfillment cash flows; we request clarification as to what is meant by trail commissions as discussed in of the 4

13 FASB ED. We question if the inclusion of trail commissions in of the FASB ED, includes any trail commissions, even those commissions that are level. Based on the examples included in and of the FASB ED, it is unclear if ultimate level commission should be included as qualifying acquisition costs or included as part of fulfillment cash flows? Question 11: Do you agree that the assumptions used in the measurement of the fulfillment cash flows should be updated each reporting period? If not, what do you recommend and why? Yes, we agree that assumptions used in the measurement of the fulfillment cash flows should be updated each reporting period. Question 12: Do you agree that the fulfillment cash flows for contracts measured using the building block approach and the liability for incurred claims for contracts measured using the premium allocation approach should be based on explicit, unbiased, and probability-weighted estimates (that is, the mean) of the future cash flows, as of the reporting date, expected to arise as the entity fulfills the contract, adjusted to reflect any contractual linkage between the contract and any underlying assets? If not, what do you recommend? We are concerned that using fulfillment cash flows (the present value of the explicit, unbiased and probability-weighted estimates of the future cash flows) as defined in the FASB ED would not include a measurement for the uncertainty related to the allocation of probable outcomes, and may not accurately reflect the measurement of the insurance liability. We believe that both the BBA and PAA should include the principle that there is uncertainty in cash flows. There is diversity among the various individuals within the AICPA group that compiled this letter regarding how the uncertainty should be included in the measurement models. There are various ways that uncertainty could be captured in both approaches such as; an explicit or implicit risk adjustment, or the amortization of the margin or premium over the coverage and settlement period. We also request further clarification as to the unbiased measurement of cash flows and the interaction with probability-weighted estimates. It could be inferred that the use of probability-weighted estimates implicitly includes the use of management bias, to determine the weightings. We recommend that the final guidance further elaborate on what judgments are allowable and still considered unbiased. Specifically we believe uncertainty is not fully included in the PAA, as the premium is amortized only over the coverage period and the liability for incurred claims is discounted. We recommend that the FASB reconsider how to include uncertainty in the cash flows. 5

14 Question 13: Do you agree with the approach in this proposed guidance to recognize changes in estimates of cash flows (other than the effect of changes in the liability arising from changes in the discount rate) in net income in the period? If not, what do you recommend? No, for the BBA we believe that changes in estimates of future cash flows, which are related to future coverages or services, should be recognized as adjustments to the margin instead of in net income. The original amount included in the margin is an estimate of future expectations, and we do not believe it is appropriate to recognize further refinements of the estimate for the future cash flows in net income. We believe recognizing estimates of future cash flows and changes to future expectations of those estimates both in the margin will result in more relevant measure of unearned profit in the financial statements. In other standards under US GAAP, information is included to help differentiate between changes in estimates related to current and future cash flows. Therefore we recommend that the FASB should consider utilizing existing guidance, and include clarification in the final guidance, for how to differentiate between changes in estimates related to current and future cash flows. Under the PAA, we request that the final guidance clarify that changes in estimated premiums (for example, changes in provisional premiums in workers compensation due to changes in head count) related to the remaining coverage should also be included as an adjustment to the liability for remaining coverage. Question 14: Do you agree that the discount rates used by the entity for nonparticipating contracts should reflect the characteristics of the insurance contract liability and not those of the assets backing that liability? Why or why not? Yes we agree that the discount rates used by the entity for nonparticipating contracts should reflect the characteristics of the insurance contract liability. Question 15: For contracts measured using the premium allocation approach, do you agree that an entity should discount the liability for incurred claims? Do you agree that entities should be allowed to elect not to discount portfolios when the incurred claims are expected to be paid within one year of the insured event? Why or why not? If not, what would recommend and why? FinREC believes that the time value of money is relevant for the measurement of liabilities. We believe that incorporating discounting into the BBA or PAA approaches should be done in conjunction with incorporating the principle that there is uncertainty related to the allocation of probable outcomes in cash flows. 6

15 As discussed in our response to Question 12, we believe that both the BBA and PAA should include the principle that there is uncertainty in cash flows. There is diversity among the various individuals within the AICPA group that compiled this letter regarding how the uncertainty should be included in the measurement models. There are various ways that uncertainty could be captured in both approaches such as; an explicit or implicit risk adjustment, or the amortization of the margin or premium over the coverage and settlement period. Specifically we believe uncertainty is not fully included in the PAA, as the premium is amortized only over the coverage period and the liability for incurred claims is discounted. We recommend that the FASB reconsider how to include uncertainty in the cash flows. We also believe that entities should be allowed to elect not to discount portfolios when substantially all the incurred claims are expected to be paid within one year of the insured event. Question 16: Do you agree that an entity should segregate the effects of underwriting performance from the effects of changes in discount rates (which would reverse over time) by recognizing changes in the present value of the fulfillment cash flows due to changes in the discount rates in other comprehensive income? If not, do you think that the effect of changes in the discount rates should be presented in net income? Please explain your reasoning. No, while we agree that an entity should separately present the effects of underwriting performance from the effects of changes in discount rates, we do not agree that changes in the present value of the fulfillment cash flows due to changes in the discount rates should be required to be included in other comprehensive income due to the potential accounting mismatches and resulting volatility in earnings. We believe that an entity should be allowed to make an accounting policy decision in an attempt to mitigate volatility in earnings (similar to the fair value option) in regards to whether changes in discount rates should be recognized in other comprehensive income or net income. We believe that this election should be consistent with any final decisions on the fair value election under the FASB s project on financial instruments. Question 17: Because this proposed Update includes the approach that changes in the insurance liability arising from changes in the discount rates should be reported in Other Comprehensive Income, do you think that a test should be required to trigger recognition in net income of some of all of the amounts in accumulated other comprehensive income (i.e., a loss recognition test based on asset-liability mismatches)? Why or why not? As noted in our response to Question 16, we believe that changes in discount rates should not be required to be reported in other comprehensive income. 7

16 No, we do not believe it would be appropriate to include a test to determine loss recognition in net income of some or all of the amounts in accumulated other comprehensive income. We believe that allowing entities to elect the appropriate accounting for changes in discount rates in conjunction with financial asset classification would help to eliminate asset-liability mismatches. Question 18: Do you agree that the method for calculating the discount rates should not be prescribed? Is the guidance on determining the discount rates understandable and operational? Are the two approaches described sufficient? If not, what do you recommend. Yes, we agree that the method for calculating the discount rate should not be prescribed. If the requirement to discount the liability for incurred claims is included in the final guidance, we believe that for simplicity, entities should have the option to elect, and make an accounting policy decision, to apply a risk free rate or other practical expedient as a discount rate for insurance contracts accounted for under the PAA. Question 19: Do you believe that interest expense generally should be based on the discount rates determined at the date the portfolio of contracts was initially recognized? Why or why not? If not, what do you recommend? We believe that the discount rate for measurement under the BBA should be the inception portfolio discount rate. We also believe that the final guidance should acknowledge that entities should have flexibility in determining the inception portfolio discount rate. Specifically, permitting an average rate for calendar year or, if elected average for a quarter. With regard to the discount rate for the liability for incurred claims under the PAA, we believe most insurers maintain their actuarial data on an incurred claim basis, and allocating IBNR amounts between policies with different inception dates would require significant costs to implement. Therefore, we recommend that entities be able to elect to either discount the liability based on the incurred discount rate or inception discount rates, as the impact of using either basis should not have a significant impact since the coverage period for most contracts under the PAA will likely be one year contracts. Margin for Contracts Measured Using the Building Block Approach Question 21: Do you agree that an insurer should not recognize a gain at initial recognition of an insurance contract (such a gain would arise when the expected present value of the cash outflows is less than the expected present value of the cash inflows), but rather should defer this amount as profit to be recognized in the future? Why or why not? Yes, we agree with the principle that an insurer should not recognize any gain at initial recognition of an insurance contract. 8

17 Question 22: Do you support using a one-margin approach, as is included in this proposed guidance, or an explicit risk adjustment and a contractual service margin (as the IASB proposes)? Please explain the reason(s) for your view. As discussed in our response to Question 12, some are concerned that using fulfillment cash flows without including a measurement for the uncertainty related to the allocation of probable outcomes, may not accurately reflect the measurement of the insurance liability. Question 23: if you support a risk adjustment and a contractual service margin, do you agree with the IASB s approach to adjust the contractual service margin for changes in estimates of cash flows? Why or why not? Do you agree with the IASB s approach to not specify acceptable approaches to determining the risk adjustment? Why or why not. Yes, we believe it is appropriate to adjust the contractual service margin for changes in estimates of cash flows which are related to future coverages or services. We agree that changes in estimates of future cash flows that do not relate to future coverage and other future services should be recognized immediately in net income. If the FASB utilizes a risk adjustment, we believe it is conceptually consistent to reflect changes in uncertainty related to future cash flows through the contractual service margin. Yes, we agree that the measurement of the risk adjustment should be based on an objective and not specific acceptable approaches, because using required approaches effectively results in a mechanical process to add an amount to the estimate. Question 24: Do you agree that a loss at initial recognition of a portfolio of insurance contracts should be recognized immediately in net income (such a loss would arise when the expected present value of the future cash outflows exceeds the expected present value of future cash inflows)? Why or why not? Yes, we agree a loss at initial recognition of a portfolio of insurance contracts should be recognized immediately. As noted in our response to Question 8, while we agree that the definition of a portfolio of insurance contracts should be addressed in the final guidance, we do not agree with the definition included in the FASB ED. We strongly recommend that the FASB and IASB converge on this issue, as differences in the definition of a portfolio of insurance contracts will result in significant reporting complications for multinational entities. The definition of a portfolio of insurance contracts as defined in the IASB ED, that takes into account how entities manage their business, is more appropriate. We also believe that for certain measurement amounts it may be necessary to group insurance contracts into a smaller unit of account than the portfolio. For example, when determining the discount rate and margin, and unlocking the margin. We recommend that the final guidance acknowledge that different groupings would be permitted. Question 25: Do you agree with the proposed method(s) of recognizing the margin (that is, as the entity is released from risk under the insurance contracts as evidenced by a reduction in the variability of cash outflows)? If not, what do you suggest and why? 9

18 We agree that it is appropriate to use a principle to determine the release from risk as evidenced by a reduction in the variability of cash outflows, but are also aware that the use of this principle may result in a lack of consistency of financial statements. We request clarification as to the intent of wording in of the FASB ED: An entity s methodology used to determine release from risk for each portfolio should be applied consistently throughout the lifecycle of the portfolio. We are unable to determine from the proposal when an entity changes its methodology used to determine release from risk because they have been able to obtain better information to determine when they are released from risk, would that change be considered a change in accounting principle or a change in estimate similar to updating information for fair value estimates? We believe such a change should be a change in estimate but the proposal can be read to imply that such a change is a change in accounting principle. We also request clarification on the wording in and of the FASB ED, related to how to do the mechanics for determining the amounts of qualifying acquisition costs to be included in the margin. We recommend that the paragraph be rewritten as the current text is confusing to apply. Acquisition Costs Question 28: Do you agree that the direct acquisition costs considered in the measurement of the margin should include only the costs directly related to the entity s selling efforts that result in obtaining the contracts in the portfolio and that all other acquisition costs should be recognized as expenses when incurred? If not, what do you recommend? We request that the FASB and the IASB strive for convergence on what qualifies as acquisition costs, as we believe it will be confusing to users of financial statements. However we believe that costs for unsuccessful efforts should not be included as acquisition costs. We also request clarification as to how changes in estimates of qualifying acquisition costs, that are considered in the measurement of the margin for contracts under the BBA or the liability for remaining coverage for contracts under the PAA, should be treated. We believe the intention of the FASB is to have any changes in estimates of qualifying acquisition costs be treated as an adjustment to the margin or the liability for remaining coverage as appropriate. However this is not clear as currently drafted, and we request it be clarified in the final guidance. As discussed in our response to Question 13, we believe that changes in estimates of future cash flows that impact qualifying acquisition costs, which are related to future coverages or services, should be recognized as adjustments to the margin or the liability for remaining coverage instead of in net income. Under the PAA, we request that the final guidance clarify that changes in estimated premiums (for example, changes in provisional premiums in workers compensation due to changes in head count) related to the remaining coverage should also be included as an adjustment to the liability for remaining coverage. 10

19 Also as discussed in our response to Question 10, we request clarification as to what is meant by trail commissions as discussed in of the FASB ED. Based on the examples included in and of the FASB ED, it is unclear if ultimate level commission can be included as qualifying acquisition costs or included as part of fulfillment cash flows? We also believe that entities should be allowed to elect to account for direct acquisition costs for contracts under the PAA, under either the proposed insurance contracts model, the proposed revenue recognition model (incremental acquisition costs) or to expense all costs. For some entities (mainly property and casualty insurance entities) the minimal benefits of including all direct acquisition costs would not outweigh the extensive costs to perform documentation of activities and the related updating of systems to capture the necessary information. Entities would be required to disclose their accounting policy election. Question 29: Do you agree that the measurement of the margin for contracts measured using the building block approach and the liability for remaining coverage for contracts measured using the premium allocation approach should be reduced for direct acquisition costs paid? If not, what do you recommend? We request that the FASB and the IASB strive for convergence on accounting for acquisition costs paid either as a reduction to the margin or included as part of fulfillment cash flows. We believe that this is a fundamental difference that will make comparability among U.S. and international insurance entities unnecessarily difficult. As noted in our response to Question 28, we also request clarification as to how changes in estimates of qualifying acquisition costs, that are considered in the measurements of the margin, should be treated. We believe it would be appropriate to have any changes in estimates of qualifying acquisition costs be treated as an adjustment to the margin or the liability for remaining coverage as appropriate. Question 30: Do you agree that an entity should recognize acquisition costs as an expense in the statement of comprehensive income in the same pattern that it recognizes the margin for contracts measured using the building block approach or in the same pattern that it reduces the liability for remaining coverage under the premium allocation approach? If not, why not? We agree that acquisition costs should be recognized as an expense in the statement of comprehensive income in the same pattern as the margin is recognized for contracts measured using the building block approach or in the same pattern that the liability for remaining coverage under the premium allocation approach is reduced. 11

20 Insurance Contract Revenue Question 31: Do you agree that users of financial statements would obtain relevant information that faithfully represents the entity s financial position and performance if, for all insurance contracts, in the statement of comprehensive income an entity presents insurance contract revenue and incurred expenses, rather than information about changes in margin (that is, the net profit)? If not, why? Although we believe it is better to present insurance contract revenue and incurred expenses in the statement of comprehensive income rather than only changes in the margin, we are concerned that the current proposed presentation approach may not provide users with a relevant financial measure. Question 32: Do you agree that, for all contracts, revenue should exclude any amounts received that an entity is obligated to pay to policyholders or their beneficiaries regardless of whether an insured event occurs, and that expenses should exclude the corresponding repayment of those amounts? If not, what do you recommend? Please specific whether your view depends on the type of contract. We agree that, for all contracts, revenue should exclude any amounts received that an entity is obligated to pay to policyholders or their beneficiaries regardless of whether an insured event occurs, and that expenses should exclude the corresponding repayment of those amounts. Question 33: For contracts measured using the premium allocation approach, do you agree that an entity should adjust the liability for remaining coverage to reflect the time value of money, and recognize the accretion of interest with insurance revenue, if the contract has a financing component that is significant to the contract? Do you agree with the practical expedient that an entity should not be required to reflect the time value of money in measuring the liability for remaining coverage (that is, if the entity expects, at contract inception, that the time period between the payment by the policyholder of all or substantially all of the premium and the entity providing the corresponding part of the coverage is one year of less)? If not, what would you recommend and why? We agree that the liability for remaining coverage should be adjusted to reflect the time value of money if the contract has a financing component that is significant to the contract. We also agree with the practical expedient that an entity should not be required to reflect the time value of money in measuring the liability for remaining coverage if the time period between the payment by the policyholder of the premium and the entity providing the coverage is one year or less. 12